Commissions, spread, market impact and a week in cash, priced on one mandate, then set against what the new manager has to add to repay them.
Replacing an investment manager costs the trading needed to move from the old portfolio to the new one plus the expected return lost while the money is out of the market, and both are paid before the new manager has earned anything. On an illustrative 400m equity mandate with 30 per cent of holdings kept in kind, the expected cost is 1,117,778, or 27.9 basis points of the mandate. A successor expected to add 25 basis points a year net repays that in 1.1 years, but at a 2.0 per cent tracking error it would take 256 years of results to show that edge is real.
Worked in full in Asset Management by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
The termination letter is free. The cost sits in four places, and only the first appears on an invoice:
The first three are often grouped as implementation shortfall. Only the traded portion bears them: securities held by both the outgoing and the incoming manager move across in kind.
| Input | Value |
|---|---|
| Mandate size | 400,000,000 |
| Overlap kept in kind | 30% |
| Commissions and taxes, on traded value | 8 bp |
| Bid-ask spread, on traded value | 12 bp |
| Market impact, on traded value | 10 bp |
| Trading days out of the market | 5 |
| Expected equity premium over cash, a year | 5.0% |
| Equity volatility, a year | 16.0% |
| Fee, old and new manager | 30 bp, 40 bp |
| New manager's expected excess return before fees, over the old | 35 bp |
| Tracking error, new manager's returns against the old | 2.0% |
Step 1, the traded value. 400,000,000 × (1 − 30%) = 280,000,000 changes hands. The 120,000,000 of overlap moves at custody cost only.
Step 2, trading cost. 8 + 12 + 10 = 30 basis points on 280,000,000:
| Component | bp | Cost |
|---|---|---|
| Commissions and taxes | 8 | 224,000 |
| Bid-ask spread | 12 | 336,000 |
| Market impact | 10 | 280,000 |
| Trading cost | 30 | 840,000 |
Step 3, opportunity cost. Five trading days in cash forgo five days of the equity premium on the traded value: 280,000,000 × 5.0% × 5 / 252 = 277,778.
Expected cost = traded value × (trading cost in bp / 10,000 + premium × days out / 252)
= 840,000 + 277,778 = 1,117,778, or 27.9 bp of the 400,000,000 mandate
Excel: =Mandate*(1-Overlap)*(SUM(CostBp)/10000+Premium*DaysOut/252)
Step 4, the risk around the expectation. The 277,778 is a mean. One standard deviation of five days' equity return is 16.0% × √(5 / 252) = 2.25 per cent, which on 280,000,000 is 6,310,485 either way. The out-of-market week is a bet the scheme did not mean to place.
Step 5, payback. The new manager is expected to beat the old by 35 basis points before fees and charges 10 more, so the net improvement is 25 basis points, 1,000,000 a year. The expected cost is repaid in 1,117,778 / 1,000,000 = 1.1 years. One standard deviation of the week in cash is 6.3 years of that improvement.
| Net improvement a year | 0 days out (840,000) | 5 days out (1,117,778) | 20 days out (1,951,111) |
|---|---|---|---|
| 5 bp, 200,000 | 4.2 | 5.6 | 9.8 |
| 10 bp, 400,000 | 2.1 | 2.8 | 4.9 |
| 25 bp, 1,000,000 | 0.8 | 1.1 | 2.0 |
| 50 bp, 2,000,000 | 0.4 | 0.6 | 1.0 |
A transition manager who keeps the portfolio invested throughout, using futures or crossing trades, removes the opportunity column entirely. Twenty days in cash, which happens when a scheme sells first and appoints later, adds 1,111,111 of expected cost against staying invested, more than doubling the payback at any improvement level, and the one-standard-deviation swing becomes 12,620,970.
The payback assumes the 25 basis points is real. To show an excess return of that size at a t-statistic of 2, with a 2.0 per cent tracking error between the two managers' returns, takes (2 × 2.0% / 0.25%)2 = 256 years of data. Even the gross 35 basis points needs 131 years. A replacement is therefore a judgement about process and people, not a reading of the next three years' numbers, and the minute recording the decision should say so. The same arithmetic on the outgoing manager is why short records prove little either way: the ratios that need the most data are set out in how to calculate tracking error and the information ratio.
Price the transition as a cash flow before the decision is taken: 1,117,778 expected on this 400,000,000 mandate, with a 6,310,485 one-standard-deviation swing from a single week in cash. Then ask what net improvement repays it and whether anyone could ever tell. The decision paper to retain or replace a manager, with the transition cost and the sentence for the minute, is among the working documents on the free companion page for this book.
It depends on how much of the portfolio has to trade and how long it is out of the market. On an illustrative 400,000,000 equity mandate with 30 per cent overlap and 30 basis points of trading cost on the rest, trading costs 840,000; five days in cash add 277,778 of expected cost, for 27.9 basis points in all.
The return the market earns while money sits in cash between selling the old portfolio and buying the new one. Five trading days on 280,000,000 at a 5.0 per cent equity premium costs 277,778 in expectation, but one standard deviation of that week is 6,310,485 either way, a risk usually removed by staying invested through futures.
Divide the expected cost by the net annual improvement after the fee difference. A manager adding 35 basis points gross at 10 more in fees adds 25 net, 1,000,000 a year on 400,000,000, and repays 1,117,778 in 1.1 years. At 5 basis points net it takes 5.6 years.
The decision to retain or replace a manager, and the transition as a cash flow, are covered in chapter 18 of Asset Management. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: corporate finance, valuation and markets → · All 1340 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.